1. Field of the Invention
The invention relates to a device applied to image extractions and, more particularly, to a device that appends a recognition mark to an extracted image so that the recognition mark can be used as a referential benchmark for image alignment when joining images.
2. Description of the Related Art
To scan an image of raw data (such as a photograph, a negative, or a slide) by a scanner or to take a picture of the data by a digital camera and make the image or the picture into an image file has become a common method for digital filing and archiving.
On the other hand, in respond to a specific scanning of a light transmitting film—for instance, a positive film, a negative film, or a slide film—a platform scanner and a film scanner have been introduced for use. Particularly, these scanners can see through an object and scan various films. Among these scanners, the film scanner is designed for scanning a 35 mm positive film and negative film. As shown in FIG. 1, the film scanner is provided with a film holder 10 that is further provided with a plurality of accommodating slots 12 for accommodating the object to be scanned. When taking a picture, first, the object 14 to be scanned is placed inside the accommodating slots 12 of the film holder 10, and then the film holder 10 will be sent into a film input device for scanning through the sides of the scanner.
However, the situation may be that the dimension of the object 14 to be scanned—for example, a photograph, a negative, or a transparent glass-slide for medical use—is larger than the dimension that a scanner can scan. For this reason, as shown in FIG. 2, the user of the scanner has to do segmented scanning on the object 14 and then join all the segmented images together. If, however, the image-joining technique is not good enough, the finished image cannot be perfect. In fact, it is very likely to get two overlapping images with poor alignment, as shown in FIG. 3.
On the other, even though software for image joining instead of the aforementioned manual joining technique can be applied, the user of the scanner still has to find a joining point for each segmented image on a computer screen to join all the segmented images. Usually, it needs to take tens of minutes to finish the joining job, followed by a polishing job that has to be done for a perfect joining. Therefore, the whole joining process is not only tedious but time consuming.
Focusing on the above-mentioned problem, the invention provides a device that appends a recognition point for image joining to the extracted image and a recognition element used by the device. Through them, a specific mark can be added into each segmented image as a referential benchmark for image alignment during the image joining.